Application of a clustering technique in identifying the “best” System of Systems (SoS) during development

This paper presents a new approach for decision making during development of a complex System of Systems (SoS) where evaluating a large number of alternatives is impractical due to resource/time limitations. We utilize a clustering technique that may greatly simplify the process of selecting the “best” SoS configuration considering multiple attributes such as capability cost, development time and systems priority. A new measure of priority is introduced in this work. The effectiveness of this method is demonstrated through a realistic case of a search-and-rescue operation, where the number of possible SoS candidates is larger than 1000. This approach provides beneficial support for decision makers during the pre-planning phase of SoS development.